In recent years, aromatic polycarbonates have been widely employed in various fields as engineering plastics which have excellent heat resistance, impact resistance and transparency. Various studies have been made with respect to processes for producing aromatic polycarbonates. Up to now processes, such as one utilizing interfacial condensation polymerization of an aromatic dihydroxy compound, such as 2,2-bis(4-hydroxyphenyl)propane (hereinafter frequently referred to as "bisphenol A"), with phosgene (hereinafter frequently referred to as the "phosgene process"), have been commercially practiced. In the phosgene process, a mixed solvent of water or an aqueous alkali solution and a water-immiscible organic solvent are generally used. Commercially, a mixed solvent of an aqueous sodium hydroxide solution and methylene chloride are employed. As a catalyst for polymerization, a tertiary amine or a quaternary ammonium compound is employed. By-produced hydrogen chloride is removed as a salt with a base.
However, in the interfacial condensation polymerization process employing phosgene, (1) toxic phosgene must be used; (2) due to the by-produced chlorine-containing compounds, such as hydrogen chloride and sodium chloride, the apparatus used is likely to be corroded; (3) it is difficult to remove impurities which adversely influence the polymer properties, such as sodium chloride, from the polymer; and (4) since methylene chloride (which is generally used as a reaction solvent) is a good solvent for polycarbonate and has a strong affinity to polycarbonate, methylene chloride inevitably remains in produced polycarbonate. Removal of the remaining methylene chloride on a commercial scale is extremely costly, and complete removal of the remaining methylene chloride from the obtained polycarbonate is almost impossible. Further, it is noted that the methylene chloride remaining in the polymer is likely to be decomposed, e.g., by heat at the time of molding, thereby forming hydrogen chloride, which not only causes corrosion of a molding machine but also lowers the quality of the polymer. Furthermore, when it is intended to produce a polycarbonate having a high molecular weight (e.g., number average molecular weight of 15,000 or more), a methylene chloride solution of such a polycarbonate has an extremely high viscosity, thereby making agitation of the solution difficult. Additionally, sticky polymer solution is produced, and hence it becomes extremely difficult to separate the polymer from methylene chloride. Therefore, commercial production of a high quality, high molecular weight polycarbonate by the phosgene process is extremely difficult.
As mentioned above, the phosgene process involves too many problems to be practiced commercially.
Meanwhile, various methods are known in which an aromatic polycarbonate is produced from an aromatic dihydroxy compound and a diaryl carbonate. For example, a process, which is generally known as a transesterification process or a melt process, is commercially practiced. In this process, a polycarbonate is produced by performing a molten-state ester exchange reaction between bisphenol A and diphenyl carbonate in the presence of a catalyst, while effecting elimination of phenol. However, in order to attain the desired polymerization degree of the final aromatic polycarbonate according to this process, phenol and, finally, diphenyl carbonate need to be distilled off from a formed molten polycarbonate of high viscosity (e.g., 8,000 to 20,000 poise at 280.degree. C.), and it is generally necessary to perform the reaction at a temperature as high as 280.degree. to 310.degree. C. in vacuo as high as 1 mm Hg or less for a period of time as long as, e.g., 4 to 5 hours. Therefore, this process has many disadvantages. For example, (1) both special apparatus (suitable for reaction at high temperatures and under high vacuum) and a special stirrer of great power (useful under the high viscosity conditions of the product to be formed) are needed; (2) due to the high viscosity of the product, when a reactor or stirring type reactor (which is usually employed in the plastic industry) is used, only a polymer having a weight average molecular weight as low as about 30,000 is obtained; (3) due to the high temperature at which the reaction is performed, branching and crosslinking of the polymer are likely to occur, thereby rendering it difficult to obtain a polymer of good quality; and (4) due to long residence time at high temperatures, discoloration of the polymer is likely to occur [see Mikio Matsukane et al, Purasuchikku Zairyo Koza 5 "Porikaboneito Jushi" (Seminar on Plastic Materials 5, "Polycarbonate Resin"), Nikkan Kogyo Shinbun Publishing Co., p. 62-67, Japan (1969)].
Moreover, with respect tot he polycarbonate obtained by the melt process, it is known that the molecular weight distribution of the polymer is broad, and that the proportion of branched structure is high. Therefore, it is recognized that the polycarbonate produced by the melt process is inferior to that produced by the phosgene process in properties, such as mechanical strength, and that, particularly, the polycarbonate produced by the melt process is disadvantageous because of its brittle fracture properties, and it is also poor in moldability because of its non-Newtonian flow behavior [see Mikio Matsukane, "Kobunshi" (High Polymer), Japan, Vol. 27, p. 521 (1978)].
Meanwhile, in the production of polyhexamethylene adipadmide (nylon 66) and polyethylene terephthalate (PET), which are examples of the most popular condensation polymerized polymers, polymerization is generally conducted by a melt polymerization process until the polymer has a molecular weight at which mechanical properties sufficient for a plastic or a fiber are exhibited. With respect to this production, it is known that the polymerization degree of the thus produced polymer can be further increased by solid-state condensation polymerization in which the polymer is heated at a temperature (at which the polymer can remain in solid-state) at a reduced pressure or atmospheric pressure under a stream of, e.g., dry nitrogen. In this polymerization, it is believed that dehydration condensation is advanced in the solid polymer by the reaction of terminal carboxyl groups with adjacent terminal amino groups or terminal hydroxyl groups. Also, in the case of polyethylene terephthalate, condensation reaction by the elimination of ethylene glycol from the formed polymer occurs to some extent simultaneously with a condensation reaction between functional groups.
The reason why the polymerization degree of nylon 66 and polyethylene terephthate can be increased by solid-state condensation polymerization is that these polymers are inherently crystalline polymers having a high melting point (e.g., 265.degree. C. and 260.degree. C.) and, hence, these polymers can remain sufficiently in solid-state at a temperature at which solid-state polymerization proceeds (e.g., 230.degree. C. to 250.degree. C.). What is more important is that, for the above-mentioned polymers, the compounds to be eliminated are substances, such as water and ethylene glycol, that have a low molecular weight and relatively low boiling point and, therefore, can readily move within and through the solid polymer so that they can be removed from the reaction system as gases.
On the other hand, it has been proposed to employ a method for producing an aromatic polyester carbonate having a high molecular weight in which a high melting temperature aromatic polyester carbonate having both an aromatic ester bond and an aromatic carbonate bond is subjected to melt polymerization, and then subjected to solid-state condensation polymerization. According to this method, an aromatic dicarboxylic acid or aromatic oxycarboxylic acid, such as naphthalene dicarboxylic acid, p-hydroxybenzoic acid or terephthalic acid, is reacted with an aromatic dihydroxy compound and a diaryl carbonate in their molten state to prepare a prepolymer. Then, the prepolymer is crystallized and subjected to solid-state condensation polymerization. If the polymerization degree is increased to some extent by melt polymerization at 260.degree. to 280.degree. C., when p-hydroxybenzoic acid is used, the resultant product is no longer in a molten state but becomes solid. Since the resultant solid is a prepolymer having high crystallinity and a high melting temperature, it is not necessary to crystallize the solid further (see Japanese Patent Application Laid-Open Specification No. 48-22593, Japanese Patent Application Laid-Open Specification No. 49-31796, U.S. Pat. No. 4,107,143, Japanese Patent Application Laid-Open Specification No. 55-98224). However, these methods apply only to the production of an aromatic polyester carbonate containing 30% or more, generally 50% or more, of ester bonds, and it has been reported that, although an aromatic polyester carbonate containing less than 30% of ester bonds was intended to be produced, fusion of a prepolymer occurred at the time of solid-state polymerization so that the solid-state condensation polymerization could not be conducted (Japanese Patent Application Laid-Open Specification No. 55-98224).
On the other hand, it is known that the presence of ester bonds as mentioned above promotes the carbonate bond-forming reaction when an aromatic polyester carbonate is produced by a melt condensation polymerization method (see Japanese Patent Application Publication Specification No. 52-36797). According to the Japanese Patent Application Publication Specification No 52-36797, when a high molecular weight aromatic polycarbonate having ester bonds is produced by melt condensation polymerization, the melt condensation polymerization reaction is markedly promoted by introducing ester bonds, in advance, into the molecular chain of an aromatic polycarbonate having a low polymerization degree. Naturally, it is believed that the above-mentioned effect of promoting the condensation polymerization reaction by the ester bonds may also be exhibited at the time of solid-state condensation polymerization. Therefore, it is relatively facile to increase the polymerization degree by solid-state condensation polymerization with respect to an inherently crystalline aromatic polyester carbonate having a high melting temperature, for example, a polymer having 40 mole % of ester bonds obtained from p-hydroxybenzoic acid, hydroquinone and diphenyl carbonate, or an aromatic polyester carbonate (such as a polymer having 55 mole % of ester bonds obtained from 2,6-naphthalene dicarboxylic acid, bisphenol A and diphenyl carbonate) which can easily become a crystalline polymer having a high melting temperature, by a simple crystallizing operation, for example, by heating at a predetermined temperature lower than the melting temperature.
However, no attempt has been made by any persons skilled in the art other than the group of the present inventors to produce a high molecular weight aromatic polycarbonate containing no ester bond by a method in which a prepolymer having a low molecular weight is first prepared by melt polymerization, and then the polymerization degree of the prepolymer is increased by solid-state condensation polymerization, except for the case where a specific highly crystalline polycarbonate having a melting temperature as high as 280.degree. C. or more has been produced by solid-state condensation polymerization (see Example 3 of Japanese Patent Application Laid-open Specification No. 52-109591). Japanese Patent Application Laid-open Specification No. 52-109591 discloses a method in which melt polymerization of an aromatic dihydroxy compound comprising about 70% of hydroquinone and about 30% of bisphenol A with diphenyl carbonate is conducted at 280.degree. C. under an extremely reduced pressure, i.e., 0.5 mm Hg to form a solidified prepolymer having a melting temperature of more than 280.degree. C., and then the polymerization degree of the prepolymer is increased by solid-state condensation polymerization at 280.degree. C. under 0.5,mm Hg for 4 hours.
However, with respect to a substantially amorphous aromatic polycarbonate comprised mainly of a dihydroxydiaryl alkane, such as bisphenol A, no noteworthy attempt has been made by any persons skilled in the art other than the group of the present inventors to produce a polymer having a high molecular weight by first forming a prepolymer having a relatively low molecular weight and then subjecting the prepolymer to solid-state condensation polymerization. For example, in the phosgene process using an acid acceptor, which is the most representative method for producing an aromatic polycarbonate, since a compound, such as sodium chloride, to be removed from the reaction system to advance the condensation reaction is generally solid in the absence of a solvent, the compound hardly moves within and through the solid polymer. Therefore, it is difficult to remove the compound from the reaction system. It is thus not feasible to carry out this method using phosgene in a solid state condensation system.
With respect to a method for producing the most popular aromatic polycarbonate, i.e., a polycarbonate derived from bisphenol A by transesterification between bisphenol A and diphenyl carbonate, all of the studies have been directed toward a melt polymerization process at high temperature under highly reduced pressure. Studies of persons skilled in the art other than the group of the present inventors have never been directed toward a method in which a prepolymer having a relatively low polymerization degree is first prepared, and then the polymerization degree of the prepolymer is increased by solid-state condensation polymerization to obtain a polycarbonate having a high molecular weight. Because polycarbonates derived from bisphenol A are amorphous polymers having a glass transition temperature (Tg) of from 149.degree. to 150.degree. C., it has been considered to be infeasible to subject polycarbonates derived from bisphenol A to solid-state condensation polymerization. In other words, in order for a prepolymer to be susceptible to solid-state condensation polymerization, it is generally required that the prepolymer not be fused but maintain its solid-state at a temperature higher than the glass transition temperature of the prepolymer (if the temperature is lower than the glass transition temperature of the prepolymer, molecular motion does not occur, thus precluding solid-state condensation polymerization). Amorphous polycarbonate which melts at a temperature of 150.degree. C. or more is practically not susceptible to solid-state condensation polymerization.
The only proposals hitherto made for producing an aromatic polycarbonate comprised mainly of a dihydroxydiaryl alkane, such as bisphenol A, which is a substantially amorphous polymer, by solid-state condensation polymerization, are those disclosed by the group of the present inventors in Japanese Patent Application Laid-Open Specifications No. 63-223035, No. 64-1725, No. 64-4617, No. 64-16826 and No. 64-16827.
Japanese Patent Application Laid-Open Specifications No. 63-223035 and No. 64-4617 disclose that solid-state condensation polymerization can be effected in the production of a polycarbonate of bisphenol A by self condensation reaction of a bisalkyl carbonate of an aromatic dihydroxy compound, e.g., bis(methyl carbonate) of bisphenol A represented by the formula: ##STR1## in which dimethyl carbonate groups are removed at an elevated temperature. In particular, in the methods of Japanese Patent Application Laid-Open Specifications No. 63-223035 and No. 64-4617, pre-polymerization is performed to obtain a prepolymer having methyl carbonate groups at both terminals thereof which is represented by the formula: ##STR2## wherein l is an integer of from 2 to about 30, the prepolymer is subjected to solvent or heating treatment for effectuating crystallization of the prepolymer, and then solid-state condensation polymerization is performed.
On the other hand, Japanese Patent Application Laid-Open Specifications No. 64-1725, No. 64-16826 and No. 64-16827 disclose that a polycarbonate of bisphenol A can be produced by reacting, for example, bis(methyl carbonate) of bisphenol A represented by formula (I) with diphenyl carbonate to produce a prepolymer having a methyl carbonate group and a phenyl carbonate group as terminal groups [such as that represented by the formula: ##STR3## wherein l is as defined above], and subjecting the prepolymer to solvent or heating treatment for crystallizing the prepolymer and then to solid-state condensation polymerization. In the methods of these patent application laid-open specifications, as different from the methods of Japanese Patent Application Laid-Open Specifications No. 63-223035 and No. 64-4617, condensation polymerization is advanced by elimination reaction of methyl phenyl carbonate from the terminal methyl carbonate and phenyl carbonate groups.
Generally, in solid-state condensation polymerization, the polymerization temperature can be low as compared to that in molten-state polymerization. Accordingly, a major advantage of a solid-state polymerization method resides in that the thermal degradation of a polymer during the polymerization step is suppressed, and that hence a high quality polymer is obtained. However, the solid-state condensation polymerization has a grave drawback in that the polymerization rate is low. In the method of producing an aromatic polycarbonate through solid-state condensation polymerization which is accompanied by the above-mentioned elimination reaction of dimethyl carbonate or methyl phenyl carbonate groups as well, the polymerization rate is not sufficiently high and hence a prolonged polymerization time has been necessary. A catalyst can be used to increase the polymerization rate in solid-state condensation polymerization. However, the catalyst is likely to remain in the final polymer, and hence the use of a catalyst is likely to cause a problem of quality degradation of final polymers (e.g., occurrence of silver streaks on the surface of a shaped article of polymers).